Ecology of fear: Ontogeny-mediated non-consumptive effects in a parasite-host system

Non-consumptive effects (NCEs) arise in the presence of parasites even when the infection does not occur and can include changes to host behaviour, physiology, or morphology. Using the Drosophila nigrospiracula – Macrocheles subbadius fly-mite system, we investigated the impact of parasite exposure (sans infection) during the pupal and adult pre-reproductive stages. First, we exposed fly pupae to mites—either indirectly (caged mites) or directly (free-roaming mites) to test the effects of parasite exposure on pupation success. Second, we tested how exposing adult female flies to mites prior to reproduction affects fecundity during the post-exposure reproductive period. We found that direct exposure to mites significantly decreased the rate of successful eclosion (development from pupa to adult) compared to unexposed pupae; however, the duration of pupation was not significantly affected. The indirect exposure did not have a significant effect on either successful eclosion or duration of pupation. We also found that indirectly exposed (caged mites) females had a significant decrease in the number of offspring produced, but only for the first few days post-eclosion, suggesting the effect was reversible after mite removal. NCEs arise after mite exposure during the pupal and pre-reproductive life stage of Drosophila, in the form of decreased eclosion success and fecundity. Investigating the NCEs associated with parasite exposure at various life stages of the host is important in understanding the ecology of fear and its total impact on hosts throughout their entire lifespan, with consequences for host ontogeny and population growth.

Methods
Pre-reproductive exposure 

(Data: Offspring.csv)

Fecundity was measured by aspirating and manually counting all F1 offspring that emerged from each of the 6 serial transfer vials. Fecundity analysis was performed using a generalized linear mixed effects model with a negative binomial distribution (glmer.nb; package lme4). The number of offspring ("Offspring") was the dependent variable and exposure ("Treatment") was the explanatory variable, with the fly mating group ("ID") as a random variable to account for natural heterogeneity between each mating group. In total, there were 18 350 offspring collected from female flies not exposed to mites (n=28) and 17 102 from those exposed to mites (n = 28).

(Data: SexRatios.csv)

A random subsample of 100 flies from each mating group ("ID") was sexed and a generalized linear model using the ‘cbind’ formula with binomial distribution (glm, family = ”binomial”) was used to analyze the effect of "Treatment" (Control = no mites; Exposed = 5 mites in a cage) on the number of "Females" versus "Males" out of a total of 100 offspring.

(Data: Weight.csv)

A subsample of 20 flies from the above-mentioned 100 sampled flies for sexing were weighed to the nearest 0.01 mg (XP105DR, Mettler Toledo, Columbus), and a generalized linear model was used to analyze the effect of "Treatment" on the "Weight" of the offspring. Mating group "ID" was included as a column, but was not found to be significant for final model selection. 

Pupae exposure 

(Data: Emerged.csv)

The effect of indirect (n= 34), direct (n= 33), or no mite (n= 36) exposure ("Treatment") on fly development was measured by counting the "Total" number of emerged adults out of 10 pupae over 7 days. Analysis was performed using a generalized linear model with a negative binomial distribution (glm.nb).

(Data: Duration.csv)

The effect of indirect (n= 34), direct (n= 33), or no mite (n= 36) exposure ("Treatment") on the "Duration" of pupal development was analyzed using a Kruskal-Wallis test (kruskal.test).